I've been thinking about this for a while. What it needs is shareable trust networks, so you can see ratings weighted by the votes of those who you particularly approve of or despise. Attempts at objective moderation just end up creating echo chambers and/or popularity contests.
I have discovered a truly marvelous proof of this, which this thread is too small to contain...but I am making notes to write it up in some useful form.
Then you'll run the risk of creating echo subchambers in a community, where people who share similar opinions form trust groups and block competing opinions out. Even if you rely on outside connections (friends and such), you'll still be more likely to network with people you agree with.
> I have discovered a truly marvelous proof of this, which this thread is too small to contain
This isn't going to be like Fermat's theorem all over again is it?
Seriously though I agree that you need a system that copes with the fact that different people have different tastes. Not a generic 'good' vs 'bad/spam'.
I have discovered a truly marvelous proof of this, which this thread is too small to contain...but I am making notes to write it up in some useful form.